Due to the severe and untreatable consequences of radiation myelopathy, many tumors in the paraspinal region receive a less than optimal radiation dose. Even a modest reduction in the risk of radiation injury from a given therapeutic dose would allow for the use of an increased dose which could enhance the probability of a local cure. The ability to accurately assess oxygen concentration in the radiated spinal cord is critical to studying the underlying mechanisms of radiation myelopathy, and in helping to determine more effective ways to use and deliver radiation to cancers in the paraspinal region. Our results show that spinal cord tissue pO2 in the irradiated rats increased significantly at intermediate (maximum level of 18 q 3 mm Hg at 55 days post irradiation) and long time periods after the radiation (average level 15 q 3 mm Hg between 90 and 180 days post irradiation), compared with the non-irradiated control group (average level of 9 q 2 mm Hg throughout the experiment). The finding that the pO2 was elevated in the irradiated spinal cord was unexpected and suggests that the prevailing expectations on the pathophysiology of radiation-induced damage to the spinal cord should be reconsidered. It seems possible that the pathogenesis is related to oxidative damage associated with the increased pO2. Although the mechanisms for the increase in pO2 in this study are not currently known, decreased oxygen consumption and/or radiation induced loss of vascular tone and microvascular remodeling and dilatation in the irradiated region are plausible factors. These results also indicate that EPR oximetry is a useful and noninvasive method for repeatedly assessing the pO2 of CNS tissue non-invasively over an extended period of time.